Abstract

Objective: To investigate whether cancer service standards are associated with survival for breast, colorectal and lung cancers at population level.

Methods: Standards of hospital cancer services in England, measured in 2001, were aggregated for 30 cancer networks covering populations of between 500 000 and 3 million people, and compared with 1-year and 5-year relative cancer survival for the incident period 1996–2001, using rank correlation.

Results: Relative survival and the cancer standards each showed statistically significant differences across cancer networks. For tumour-specific services, the total score of 35 standards was associated with longer relative survival for both colorectal and lung cancers (p<0.05), but not breast cancer, while colorectal cancer survival was strongly (p<0.01) associated with the specific standard “written agreement describing referral guidelines”, and lung cancer (p<0.05) with two other guideline standards. There were also associations of longer survival with two measures of nursing staff specialist qualifications. Compliance with general standards for cancer services was not associated with survival for breast cancer, and showed only borderline (p<0.1) associations for colo-rectal cancer, while some standards on medical and management lead staff were significantly associated (p<0.05) with poorer survival for lung cancer. Overall, compliance with standards for hospital pathology and radiology services also showed no associations with survival.

Conclusion: This study suggests that compliance with some clinical service standards, such as guidelines, could contribute to better survival at population level, while more general organisational aspects of cancer services may not directly improve survival.